The Social Impact Bond is a topical, a la mode form of payment by results (PbR) device, to enable social organisations to deliver services in the public sector.
This short article draws on detail a paper by the Charities Aid Foundation (CAF), Funding Good Outcomes, from the Autumn of 2012 and from a more recent article by Caroline Fiennes, in an issue of Philanthropy Impact Magazine, What the First Social Impact Bond Won’t Tell Us.
The CAF paper makes a good case for PbR, and how with support the social sector can engage with commissioning bodies and, with its focus on outcomes, use the sector’s sensibility and sensitivity to change in a community of interest to achieve potentially profound results.
PbR based on outcomes presents a real opportunity for not for profit organisations to win public service delivery contracts. As the focus is shifted away from the exact nature of the service towards the outcomes produced, there is more room for innovation and greater freedom for not for profits to demonstrate effectiveness in their approach.
The CAF paper looks at several examples of new, or newish, PbR programmes, from The Work Programme, The Youth Contract and the Peterborough Pilot. The paper does recognise the singular and distinctive flaw in some PbR provision for the social sector. Risk, and the inherent pressure of working capital needs as the service is rolled out to accommodate the strictures of the commissioned work.
CAF cite examples where PbR has driven sector players to extinction as a result of this pressure. They call upon the social finance sector to look at alternative methods of capital deployment to accommodate these new service delivery models.
They cite the need for Social Finance to lend to PbR contractors, in order to help sustain the emerging contract.
They call for options which see the commissioning body contracting with the Social Finance intermediary – with the Social Investment Financial Intermediary (SIFI) mitigating risk by calling upon a spread of external social investors to support the contract delivery.
Thay call for SIFI’s to act as underwriters of the contract, providing payment guarantees to well monitored and managed programmes of work. The interesting example they use is the Goldman Sachs model, where a $10 million dollar Social Impact Bond was funded to transform outcomes at Rikers Island Prison in New York.
So, even within the development of new Social Finance modalities, there is the opportunity to be subtle and agile in approaches to contract structure, whatever the scale of the investment, this author would argue.
In the article by Caroline Fiennes, What the First Social Impact Bond Won’t Tell Us, the author looks at the prison system too. This time in Peterborough in the UK.
Fiennes presents an argument, that even with elaborate arrangements in place and the agreement to deliver, the analysis of performance to assess the PbR outcomes can still be a flawed and inherently mis-leading process.
The Peterborough Pilot formal agreement sets out processes to repay investors when the agreed outcomes are reached. So far so good. However, Fiennes argues that assessing the level of re-offending by a target group of prisoners, which the contract uses to trigger repayments, is essentially based on flawed statistical assumptions.
The control group in Peterborough and the individuals receiving the ‘treatment’ are correlated by using a system called Propensity Score Matching. In this case the PSM is of a particularly elaborate kind. Using ten ‘control’ prisoners for each ‘treatment’ prisoner.
Fiennes argues that this methodology…
only ever looks at indicators which are observable, such as age, background and criminal history. Yet is often unobservable factors – such as attitude or resilience – that drive behaviour.
Secondly, the data used is stored on the Police National Computer, which itself is of a very basic nature…where it cannot distinguish whether somebody had problems or a history of heroin use, which obviously would influence their behaviour and the care they need.
The Fiennes paper also argues that it is going to be difficult, even with a ‘successful’ contract outcome, to assess the comparative strengths of the Bond programme, when set against, for example, the good an enlightened and responsive prison governor might achieve, even without the Bond.
Professor Sheila Bird of Cambridge University opines that all of these problems could have been averted, if the the first Social Investment Bond in the UK had been tested against a known intervention with a conventional funding mechanism.
So even having succumbed to the lure of the new, there is much complex reflection needed to justify these new finance tools and to successfully measure their outcomes appropriately.
This does not, we would argue, negate their importance or cease to offer SIFI’s and commissioning bodies examples of financing social outcome, even in areas and communities where the scale of investment and the outcome expected is of more modest proportions.